The literature on the relationship between government debt and economic growth deals extensively about the direction of the causation between these two macroeconomic aggregates. Apart from the causation that runs from government action to GDP, a decrease of GDP can also lead to an increase in government debt – the reversed causation effect. In recessionary times GDP may shrink and, as a result, government revenues go down. This causes government debt to rise as long as the government does not change the stance of fiscal policy. Therefore, an ailing economy can cause government debt to rise and not the other way round. Other authors hold the view that a non-linear relationship between government debt and economic growth exists. With increasing government debt the positive impact of the accompanying public spending increase is compensated by adverse consequence of a crowding-out of private investments and a rising uncertainty of future fiscal policy.

Although so far empirical evidence has been mixed, a majority of studies find either a negative correlation or a non-linear relationship between government debt and growth. In addition, a few studies analysis the causal relationship and come to the conclusion that slow growth leads to a run-up of public debt.

This paper contributes to the literature on the debt-growth-nexus by analyzing a large historical panel dataset of 17 OECD countries that ranges from 1870 to 2016. For this we apply a two-way-fixed-effects model with robust standard errors. Our estimations suggest that the correlation between government debt and growth is non-negative. While our baseline regressions support the view that government debt is harmful for economic growth, this relationship disappears and loses statistical significance after a number of robustness tests such as splitting the sample into a pre- and post-world-war-II period, applying dummies for financial crises or including country-specific trends. Controlling for one of these factors, the result that government debt is associated with decreasing economic growth - that has been found in a number of recent studies, as well as in our baseline specification - does not hold. Rather than suggesting that government debt is at the roots of slow growth, our evidence indicates that slow growth brings about an increase of public debt.